Current and historical trends in general aviation in the United States

Current and Historical Trends in General Aviation

in the United States

by

Kamala Irene Shetty

B.S., University of Maryland (2010)

Submitted to the Department of Aeronautics and Astronautics

in partial fulfillment of the requirements for the degree of

Master of Science in Aeronautics and Astronautics

at the

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

September 2012

c

Massachusetts Institute of Technology 2012. All rights reserved.

Author . . . .

Department of Aeronautics and Astronautics

August 23, 2012

Certified by . . . .

R. John Hansman

Professor of Aeronautics and Astronautics

Thesis Supervisor

Accepted by . . . .

Eytan H. Modiano

Professor of Aeronautics and Astronautics

Chair, Graduate Program Committee

Current and Historical Trends in General Aviation in the

United States

by

Kamala Irene Shetty

Submitted to the Department of Aeronautics and Astronautics on August 23, 2012, in partial fulfillment of the

requirements for the degree of

Master of Science in Aeronautics and Astronautics

Abstract

General aviation (GA) is an important component of aviation in the United States. In 2011, general aviation and air taxi operations represented 63% of all towered opera-tions in the United States, while commercial aviation was responsible for 34% of those operations. It is clear that GA is a considerable component of the national airspace and airport system, even when only accounting for towered operations. Because of this significant presence, insight into GA is relevant to issues in air traffic manage-ment, air transportation infrastructure, and aviation safety, among others. Beyond the operational aspect, GA is of significance to society as a whole and to other stake-holders, including pilots groups, aircraft manufacturers, and the work force. In 2009, general aviation generated 496,000 jobs and its total economic contribution to the U.S. economy was valued at $76.5 billion.

However, a comparison of general aviation’s impact on jobs and on the economy between 2008 and 2009, shows a 20% decrease in jobs and a 21% decrease in total economic impact in the course of a year. There is also a significant decreasing trend in the active pilot population, along with steady decreases in GA flight hours and towered operations.

The objective of this thesis is to explore the details of these changing trends and to determine what drives and what hinders general aviation activity in the country. A combination of data analysis and the development of a survey administered to general aviation pilots shed light on what has driven activity in the past on a national scale, what factors affect an individual pilot’s level of activity, and what challenges the general aviation community faces in the future.

Thesis Supervisor: R. John Hansman

Acknowledgments

This research was made possible through funding from the Federal Aviation Admin-istration.

The help from general aviation publications, including AVweb, has had a signifi-cant role in the success of this project. Of course, much gratitude is also extended to the generous response from the general aviation community and to the hundreds of pilots that took the time to participate in this project.

I would like to thank Professor Hansman for his guidance, feedback, and under-standing through this research project and I am grateful to have learned just from just a fraction of his extensive expertise. I also thank my lab mates in the Inter-national Center for Air Transportation along with Thea Graham, Dan Murphy, and many others in AGJ-6 at the FAA for providing a great source of help and support.

A special thanks to Mom, Dad, Justin, and Khashayar for the love and support that only family can provide. I dedicate this work to Grandpa, who has always been a source of inspiration.

Contents

1 Introduction 15

1.1 Research Motivation . . . 15

1.1.1 What is General Aviation . . . 15

1.1.2 Current Trends in General Aviation . . . 16

1.1.3 General Aviation Stakeholders . . . 18

1.2 Thesis Overview . . . 20

2 General Aviation Forecasts 21 2.1 Types of Forecasts . . . 22

2.1.1 FAA Aerospace Forecast . . . 22

2.1.2 FAA Terminal Area Forecast . . . 22

2.2 Analysis of 2006 TAF Forecast . . . 24

2.2.1 All Airports . . . 24

2.2.2 Towered Airports . . . 28

2.3 2001 - 2011 Aerospace Forecasts . . . 30

2.3.1 Comparison of Past Forecasts . . . 30

2.3.2 2011-2031 Forecast . . . 30

2.4 Summary . . . 31

3 Trends 33 3.1 Activity Trends . . . 33

3.1.1 Operations Trends . . . 33

3.1.2 Hours Flown Trends . . . 35

3.1.3 Spatial Trends . . . 35

3.2 Drivers and Related Trends . . . 36

3.2.1 Pilot Population . . . 36 3.2.2 Aircraft Population . . . 38 3.2.3 Fuel . . . 41 3.2.4 Socio-economic factors . . . 43 3.2.5 World Events . . . 45 3.2.6 Safety . . . 45 3.3 Conclusions . . . 47

4 Survey 49 4.1 Goals . . . 49 4.2 Design . . . 50 4.2.1 Survey type . . . 50 4.3 Sampling Method . . . 52 4.4 Sources of Error . . . 52 4.5 Response Profile . . . 54

4.5.1 Certifications and Ratings . . . 54

4.5.2 Age . . . 55 4.5.3 Gender . . . 56 4.5.4 Regions . . . 57 4.5.5 Occupation . . . 57 4.6 Summary . . . 58 5 Survey Results 59 5.1 Flight Hours . . . 59 5.2 Flying Beginnings . . . 60 5.3 2011 Activity . . . 62 5.3.1 Not Active . . . 62 5.3.2 Active . . . 63 5.4 Peak Activity . . . 64 5.5 Opinions . . . 68 5.6 Future Outlook . . . 70 5.6.1 Levels of Activity . . . 70 5.6.2 Fuel . . . 71 5.6.3 Activity Stimuli . . . 71 5.6.4 Current Challenges . . . 73 6 Conclusion 75

A Survey of Active and Inactive General Aviation Pilots on Factors

List of Figures

1-1 2009 General Aviation and Part 135 Survey Results: Total Hours

Flown by Use [2] . . . 16

1-2 Historical Trend of Commercial and General Aviation [4] . . . 16

1-3 GA activity and GDP [4] . . . 17

1-4 2011 National Plan of Integrated Airport Systems (Image Source:[6]) 18 2-1 Commercial Growth Forecast [14] . . . 25

2-2 General Aviation Growth Forecast [14] . . . 26

2-3 Comparison of GA and Commercial Projected Growths [14] . . . 26

2-4 2006 Commercial Operations Forecast [14] . . . 27

2-5 2006 General Aviation Itinerant Operations Forecast [14] . . . 27

2-6 2006 General Aviation Local Operations Forecast [14] . . . 28

2-7 2006 Towered Commercial Operations Forecast [14][4] . . . 29

2-8 2006 Towered General Aviation Itinerant Operations Forecast [14][4] . 29 2-9 2006 Towered General Aviation Local Operations Forecast [14][4] . . 29

2-10 Local General Aviation Operations at All Towered Airports [15] . . . 30

2-11 Itinerant General Aviation Operations at All Towered Airports [15] . 31 3-1 Towered Operations[4] . . . 34

3-2 Hours Flown by Use [2] . . . 34

3-3 General Aviation Airports and Weather [4] . . . 35

3-4 Pilot Certificates Held by Category [16] . . . 36

3-5 Percentage of Pilot Certificates Held in the U.S. Population, by Cate-gory [16] . . . 37

3-6 Instructional Hours [2] . . . 37

3-7 Average Age of Active Pilots by Category [16] . . . 38

3-8 General Aviation Shipments by Type of Aircraft Manufactured in the United States [16] . . . 38

3-9 Towered General Aviation Operations and Events Affecting Aircraft Population . . . 39

3-10 General Aviation Total Shipments, Less Foreign Exports, by Type of Aircraft Manufactured in the United States [16] . . . 40

3-11 General Aviation Total Billings by Type of Aircraft Manufactured in the United States [16] . . . 40

3-12 Prime Supplier Sale Volumes [17] . . . 41

3-14 Towered General Aviation Operations and Rising Fuel Prices . . . 42

3-15 Disposable Personal Income Per Capita Chained 2005 Dollars [21] . . 43

3-16 Unemployment [22] . . . 44

3-17 Towered General Aviation Operations and Economic Recessions . . . 44

3-18 Towered General Aviation Operations and World Events . . . 45

3-19 General Aviation Accidents [26] . . . 46

3-20 General Aviation Fatalities [26] . . . 46

3-21 Factors Affecting General Aviation . . . 47

4-1 Survey Respondents’ Certification vs. FAA Active Airmen Certifica-tion Estimates . . . 55

4-2 Survey Respondents’ Age Distribution vs. FAA Total Active Airmen Age Distribution Estimates [28] . . . 56

4-3 States in which Respondents were Primarily Based in 2011 . . . 57

4-4 Regions in which Respondents Flew in 2011 . . . 58

5-1 Survey Results - “What is your total flight time?” . . . 59

5-2 Survey Results - “When did you start flying?” . . . 60

5-3 Survey Results - “There are many reasons why someone makes the decision to learn to fly. Please select all of the following that applied to you.” . . . 61

5-4 Survey Results - Reasons for respondents began flying by years they began flying . . . 61

5-5 Survey Results - “Please explain why you did not fly in 2011?” . . . . 62

5-6 Survey Results - “Approximately, how many hours did you fly in 2011?” 63 5-7 Survey Results - “What type of aircraft did you fly the most during 2011?” . . . 63

5-8 Survey Results - “During what time period was your flying activity at its highest?” . . . 64

5-9 Survey Results - “Please explain why 2010-2011 was the time of your highest flying activity?” . . . 65

5-10 Survey Results - “Approximately, how many hours did you fly in a year during your period of peak flying activity?” . . . 65

5-11 Survey Results - “What type of aircraft did you fly the most during your period of peak flying activity?” . . . 66

5-12 Survey Results - “What impact did the change in the following factors have on your flying between 2011 and your peak flying years?” . . . . 67

5-13 Survey Results - “To what extent do you agree or disagree with the following statements?” . . . 69

5-14 Survey Results - “How do you expect your level of flying activity to change over the next few years? ” . . . 70

5-15 Survey Results - “How do you expect your level of flying activity to change over the next few years? ” . . . 71

5-16 Survey Results - “Is your flying activity dependent on fuel costs? At what fuel price (per gallon) would your flying activity significantly

decrease or stop completely?” . . . 72

List of Tables

1.1 Number of Airports in the NPIAS by Category . . . 19

3.1 Economic Recessions in the United States [23] . . . 44

Chapter 1

Introduction

1.1

Research Motivation

The goal of this study is to gain a better understanding of the factors that drive general aviation activity in the United States. Historical levels of activity are analyzed along with changes in related time series that include fuel costs, aircraft shipments and average ages, numbers of active pilots, socioeconomic indicators, etc. Another aspect of the study involves a survey of general aviation pilots in order to gain additional insights into these factors by gathering details from the experience of individual pilots and the reasons they choose to fly or to stop flying. The results of the survey along with the analysis of historical trends will help identify any challenges facing the growth of general aviation and its role in the greater aviation community, including what may promote or hinder activity in the future.

1.1.1

What is General Aviation

Commercial aviation activities in the US are easily defined as all operations that are regulated by Part 121 of the Federal Aviation Regulations put forth by the FAA, which encompasses all scheduled air carrier service [1].

The term general aviation, on the other hand, is a catch-all phrase for all aviation activities that doe not fall under commercial aviation, major cargo or military opera-tions, and covers a broad spectrum of aviation activity uses. The FAA, in its annual General Aviation and Part 135 Survey, categorizes GA activity by use. Figure 1-1 shows the different categories of GA activity and their corresponding portion of all hours flown in 2009.

From an operational standpoint, these GA activities can be categorized as either local or itinerant operations, whereas commercial aviation is almost exclusively itin-erant. Operations are defined as either an arrival or departure of local or itinerant nature. Local operations are defined as operations performed by aircraft that are operating within the local traffic pattern of the airport or within sight of the airport, are going to designated local practice areas within 20 miles of the airport, or are performing simulated instrument approaches or low passes at the airport [3]. For instance many operations that fall within the personal or instructional use category

Figure 1-1: 2009 General Aviation and Part 135 Survey Results: Total Hours Flown by Use [2]

are categorized as local. Itinerant operations are defined as the compliment of local operations, i.e. aircraft coming from or going to a different airport. Most business or corporate transportation uses would call under itinerant operations.

1.1.2

Current Trends in General Aviation

Figure 1-2 compares historical trends for commercial operations and general aviation

operations at towered airports across the country. The plot shows that the two

categories have not followed the same general trends and are driven by different factors.

Historical trends have shown that the growth in annual commercial air travel, measured in revenue passenger kilometers (RPK), is approximately twice the annual growth rates of gross domestic product (GDP) [5]. This strong relationship to GDP indicates a significant dependence on the national and world economy and commercial operations, through some relationship with RPKs, should change in a tractable way. Unlike commercial aviation, GA in recent history has not closely followed aggregate economic indicators, such as the GDP. This is illustrated in Figure 1-3. Even in the decades before 2000, when activity on an average trended upwards along with growth in GDP, there were significant anomalies that indicate that there are other major factors at play.

Figure 1-3: GA activity and GDP [4]

Since GA is such a broad category of aviation, many differing drivers are working at one time to affect aggregate levels of operations. Some of these factors include economics, the volatility of fuel prices, increased use of internet in business (hence, decreasing business travel), tax incentives for aircraft ownership, the costs of owning and operating personal aircraft, the total private pilot and GA aircraft populations, and many more.

Two things are now clearly evident. First, it is clear that while commercial avi-ation and general aviavi-ation may share some drivers, such as economic growth, they do have their own distinct characteristics and factors affecting their levels of activity. This leads to the question of whether or not forecasting one group over the other has shown to be more accurate. Second, it is clear that general aviation activity at an aggregate level has suffered a significant and seemingly persistent decline over the past 12 years.

1.1.3

General Aviation Stakeholders

Many groups are interested in and play a role, in one way or another, in general aviation and its future in the United States.

Airports

The FAA’s National Plan of Integrated Airport Systems (NPIAS) identifies all domes-tic, public use airports that are considered a significant part of the civil air transporta-tion system. As of February 2010, There were a total of 19,734 airports (including heliports and seaplane bases), of which only 3,380 are included in the NPIAS [6]. Airports that are included are categorized as follows [6]:

• Primary Commercial Service Airports - airports receiving scheduled passenger service and having 10,000 or more enplaned passengers per year

• Nonprimary Commercial Service Airports - airports receiving scheduled passen-ger service and having 2,500 or more enplaned passenpassen-gers per year

• Reliever Airports - airports that relieve congestion at Commercial Service Air-ports by providing improved general aviation service

• General Aviation Airports - all other airports

Figure 1-4 maps out all airports by its category. It is clear that there is a significant amount of airports that solely accommodate GA activity as compared to the number of primary and secondary airports which, in general, are also able to accomodate GA activity. Table 1.1 shows the actual number of airports by category and their corresponding percentage of the NPIAS. Commercial aviation is concentrated in about 15% of the nation’s significant airports, while GA activity is clearly more dispersed.

Table 1.1: Number of Airports in the NPIAS by Category

Primary 382 11%

Nonprimary 121 4%

Reliever 269 8%

General Aviation 2,560 77%

The type of activity at an airport obviously affects the services that an airport provides, ranging from runway sizes to hangar space to passenger processing areas and so on. It is, therefore, also obvious the need to distinguish between types of operations when planning and operating an airport. The greater the understanding of how general aviation is evolving across the country allows airports to better plan for today and for the future.

Government

Along with local planners and governments, the federal government is very interested in the demand for airports, as it can base its decision on funding of airport projects on the expected utilization of their investments. The FAA is charged with the task of distributing grants across airports in the NPIAS so that the net benefit to the whole air transportation network and society as a whole is maximized. Those grants are distributed through the Airport Improvement Program (AIP), a pool of money that is funded through the collection of airline ticket taxes on all commercial flights [7].

In 2009, $3.47 billion was distributed through 2,885 grants across the country [8]. Only 30.7% of the funds were given to the projects at the largest commercial airports, where a disproportionately larger amount of funds, (88.6%) are collected from all passenger enplanements. Other airports at which significant GA activity is present along with commercial activity collects 65.5% of those funds. The smallest of all airports, those designated as general aviation airports, receive the greatest amount of grants, 52.9%, and receive a significant 17.3% of the funds.

In addition to the importance at the airports, the FAA is interested in general aviation activities in planning for future workload at air traffic control towers and for the management of the National Airspace System (NAS).

American Public

Whether or not its importance is fully appreciated by society, general aviation pro-vides importance service to the American people. General aviation includes activities that are crucial to small and large business, agriculture, law enforcement, flight train-ing, medical evacuation, tourism, and travel to remote areas.

General aviation also contributes to the American economy. In 2009, general avi-ation generated 496,000 jobs and its total economic contribution to the U.S. economy was valued at $76.5 billion [9]. However, a comparison of general aviation’s impact

on jobs and on the economy between 2008 and 2009, shows a 20% decrease in jobs and a 21% decrease in total economic impact in the course of a year.

Aircraft Manufacturers

A significant portion of the economic impact of general aviation, 33.7% in 2009, comes from aircraft manufacturing. The aircraft manufacturing business relies on the levels of general aviation activity in the country and needs to know its future in order to conduct business efficiently.

The general aviation fleet clearly dominates commercial aircraft in terms of num-bers. In 2008, 96% of the active fleet in the United States were general aviation aircraft with 228,662 aircraft versus 7,856 in the country’s commercial fleet [10].

1.2

Thesis Overview

Given the current state of general aviation in the United States and the number of its unique stakeholders, a greater understanding of its drivers is of significant importance. The goal of this study is to gain a better understanding of the trends involving general aviation, including trends in its activity levels, drivers, and forecasting, as well as current attitudes and outlooks.

Chapter 2 overviews the major forecasts used in the US for general aviation ac-tivity. It examines old forecasts and analyses the accuracy of those forecasts and compares them to those of commercial aviation, in order to determine whether there is a gap in forecast error between the two groups. GA forecasts are described in general terms, along with a short summary of the FAA’s 2011 forecast for general aviation over the next 20 years.

Chapter 3 is a survey of the historical and current trends in general aviation. It first looks at different measures of activity, including towered activity from the FAA’s ATADS and hours flown data from the FAA’s General Aviation and Part 135 Survey. It then goes on to characterize the factors affecting this activity including historical trends of fuel, socio-economic factors, pilot population, aircraft population, safety issues, and some other exogenous factors than may have an effect on national levels of activity.

In Chapter 4, a sample survey of general aviation pilots is introduced as a way to supplement the data analysis of Chapter 3 in understanding what drives activity levels. The details of the survey design and methods are discussed, along with any issues that may affect the interpretation of the survey results.

The details of the survey responses are presented in Chapter 5. An analysis

is prepared at the aggregate level and then again within subgroups of the survey respondents.

Chapter 2

General Aviation Forecasts

Chapter 1 discussed the importance of general aviation to society and to its various stakeholders. Many, if not all, of those stakeholders are interested in the future of this subset of aviation, and many either utilize or develop forecasts of GA activity or demand to help make decisions, so as to prepare for the future or to makes changes to alter the course that it may take.

This chapter is not meant to be an in-depth study of general aviation forecasting or of aviation forecasting in general. The goal, however, is two fold. The first goal is to help bolster the argument that further research into the drivers of general aviation activity is indeed warranted. The second is to look at and characterize the current forecasts for general aviation activity in the United States.

One of the main practical goals of identifying the drivers of general aviation is to use that information in the future. As with any phenomena, the forecaster’s job of predicting the future is made easier when the system can be modelled as accurately as possible. While a forecast’s realized inaccuracy may not necessarily indicate a lack of understanding of the phenomena, it can be the result of difficulties in predicting driving forces in a necessarily unpredictable world, an incomplete model of the process should most definitely have a negative impact on the accuracy. Therefore, analyzing the accuracy of general aviation forecasts, especially in comparison to other forecasts, can provide evidence bolstering the argument for further investigation of general

aviation. To do this, the results of the FAA’s 2006 Terminal Area Forecast are

analyzed and the accuracy of the short term forecasts for GA and commercial activity are compared.

The last part of the chapter looks at the the FAA’s Aerospace Forecast released in years 2001 to 2011. A comparison of the forecasts for the last 10 years helps characterize the general outlook on general aviation as seen by forecaster’s at the FAA. In later chapters, trends and and other qualitative data will be compared against the current forecast, in order to help qualitatively judge the reasonableness of the forecast.

2.1

Types of Forecasts

Forecasts of general aviation can include forecasts of hours flown, operations, total fleet, based aircraft or any other metric. They can be focused on aggregate levels across the nation or be focused on the regional, state or airport levels.

While airports, regional aviation authorities, and other stakeholder create fore-casts of GA activity ,the two biggest forefore-casts in the United States,the FAA’s Aerospace Forecast and Terminal Area Forecast, are discussed below.

2.1.1

FAA Aerospace Forecast

The FAA Aerospace Forecast is a yearly report that presents a 20 year outlook on commercial and general aviation at the national level.

For general aviation, the reports includes forecasts for: • Active aircraft

• Hours flown

• Active pilots by type of certification • Aircraft fuel consumption

• Towered (local and itinerant) operations

The methodologies used to forecast these metrics are not explicitly described, however, the following points are made throughout the report and give some indication on the forecast methods:

• Growth rates are applied to current estimates of fleet size, hours flown and utilization from the General Aviation and Part 135 Activity Survey [2].

• The growth rate for the fleet is based on assumptions on fleet attrition in conjunction with General Aircraft Manufacturers Association (GAMA) aircraft shipment statistics.

• The growth rate of flight hours are based on assumptions on utilization in conjunction with the forecasted fleet.

• The forecast is formulated in a similar fashion year to year in hopes the error of the forecast in relation to realized levels can be estimated from year to year.

2.1.2

FAA Terminal Area Forecast

The FAA’s Terminal Area Forecast (TAF) is an official activity forecast for all air-ports in the NPIAS and is used by many groups in the federal government, state governments, local airport authorities, and aviation operators for planning. The TAF is based on the results of the Aerospace Forecast.

• total enplanements for air carriers and regional carriers

• total itinerant operations for air carrier, air taxi and commuter, general aviation, and military activity

• total local operations for general aviation and military activity • total instrument operations

The methodology used to create the commercial and GA operations forecasts in the TAF is abstractly outlined [11] and summarized below:

• The TAF is an unconstrained forecast, meaning the forecast is independent of current capacities of airports or the national airspace system and is dependent only on the driving factors of aviation demand.

• Forecasts are formulated from historical relationships between activity and other local and national factors and are compared to an airport’s historical trend. Regression analysis and other growth rates are used as well. No specific models are identified.

• The historical data for activity at airports with tower service can be assumed to have negligible measurement error. For all other airports (a majority of GA airports), operations are taken from FAA Form 5010, which are estimates given by inspectors, or from other various sources.

• Thirty five airports that are included in the FAA’s Operation Evolution Plan (OEP), large commercial hubs of special significance to the air transportation system [12], are subject to a more thorough analysis which takes into account:

– local socio-economic factors

– growth rates of origin-destination and connecting passengers – prices into the airport

– trends in load factors and aircraft sizes

– future schedules from the Official Airline Guide (OAG)

• The forecast is constructed by the Office of Aviation Policy and Plans, sent out to FAA regional offices for reviews and suggestions are taken into consideration. The forecast also may utilize forecasts from local airport authorities or airport master plans, if the methods of forecasting are approved by the FAA.

The TAF is an important source of information for policy makers and planners, but it is clear that it has its limitations, and should be supplemented with more in depth analysis.

The documents accompanying the TAF for public release do not stress the meth-ods used for non-towered airports. Also, the TAF presents a single forecasted value to the public, with no indication of confidence in those values. It does not give any

level of confidence in the presented forecast, nor does it make any other attempt to show a range of values the future could see(i.e. giving optimistic and pessimistic assessments). This may be due to political reasons or due to the fact that the TAF is utilized by many different parties, some of whom do not have in depth knowledge of forecasting issues.

Some smaller issues with the methodologies used to formulate the TAF include the fact that the TAF uses inputs from airport master plans and local airport authorities. This is a double edged sword. While local officials have a better intuition of where operations trends are headed and are more familiar with the issues facing the airport that can affect activity, it can also lead to bias towards optimism. One source of this bias is the competition within the AIP funding program itself, as airports will have to justify that its proposed projects are worthy of grants with levels of future activity worthy of the investments.

Another issue with the TAF and many other aviation forecasts is that since the forecasts are based on historical levels of activity, which can be constrained, regressing on the historical values may not reveal the true unconstrained level of demand.

2.2

Analysis of 2006 TAF Forecast

The 2006 forecast was chosen for this analysis, since it was the oldest forecast available for public query. For the purposes of studying the general accuracies of these forecasts, a forecast model that contains more than four realized time periods would have been ideal. This is because short term conditions (i.e. four years into the future) will most likely follow the levels and growth rates prevailing at the time of the forecast and then may be less of an indication of the underlying assumptions of the forecasting methods [13]. Unfortunately, the analysis is dictated by the information currently readily available to the public.

2.2.1

All Airports

The model has forecasts for 3,392 facilities and queries to this model will return with historical activity levels from fiscal years 2001 - 2006 as well as forecasted values at every year between fiscal years 2007 and 2025. The following analysis first looks at the national forecast and then looks at the characteristics of the towered airports separately.

Characteristics of Growth

Of the 3,392 facilities in the forecast, 59% served non-zero levels of commercial ac-tivity in 2006. It was also noted that no airports in the system were forecasted to accommodate commercial activity that did not do so previously. The types of growth for these airports, the percentage change from the base level in 2006 to projected levels in 2025, is shown in Figure 2-1. The most striking observation is the fact that the majority of airports are assumed to have no change in commercial operations. Of

those with projections for change from the 2006 levels, the majority were assumed to have positive growth in operations through 2025. Figure 2-1 also shows the relative frequency of those airports that were given positive or negative forecasts by the per-centage growth they are assumed to have in the TAF. The distribution is somewhat of a bell curve, with a median of 22% growth over the two decades. There were a few outliers, that had growth in the thousands of percent. These outliers are the airports that were projected to serve an increased number of operations from a very low base level.

Figure 2-1: Commercial Growth Forecast [14]

Similarly, no airports in the system were forecast to accommodate general aviation that did not do so in 2006. Given that the portion of the NPIAS that serves general aviation is 95.6% in the baseline year, there isn’t a large potential opportunity to spread. The expected growths for these airports that serve GA is summarized in Figure 2-2. The amount of airports with forecasted change in GA operations was only 25%, compared to the 32% for the same metric for commercial operations. Figure 2-2 also shows the histogram of non-zero growths for GA activity. It is interesting to note that the variability in growth is smaller than that for the same distribution for commercial activity and that the median of the distribution is slightly higher (29% vs. 22%).

Figure 2-3 corresponds to the subset of the NPIAS that had both commercial and GA operations. The forecast continues to expect that these 611 airports will serve both types of activity. The histogram tallies how GA percentage growth and commercial growth are related at a specific airport, for all the airports. The horizontal axis is the range of multipliers k, where commercial growth = k ∗ GA growth. The distribution is centered around zero, indicating that at a majority of airports, assumed growth rates of commercial and GA operations were approximately the same.

National Forecast

The 2006 TAF projects a 40.9% increase in nationwide commercial operations, while GA operations are projected to grow by 17.3%.

Figure 2-2: General Aviation Growth Forecast [14]

Figure 2-3: Comparison of GA and Commercial Projected Growths [14]

Figures 2-4, 2-5, and 2-6 plot the national forecasts for commercial operations,

itinerant GA operations, and local GA operations. Along with the forecast, the

historical data for each category of operations is plotted. The historical values from the 2006 TAF are plotted in green, while the historical values from the more recent 2010 TAF are plotted in red. The dotted line shows a best-fit curve of the forecast values. The insert boxes zoom into the realized and forecasted operations levels for the years 2007 to 2010, where above each year is the absolute difference and the

percentage difference (realized−f orecasted_realized ) between the two curves.

All three categories of operations decreased in the 4 year period, while the corre-sponding forecasts have shown increasing levels of activity. The forecast for national commercial activity followed an exponential curve virtually perfectly, while GA

fore-Figure 2-4: 2006 Commercial Operations Forecast [14]

Figure 2-5: 2006 General Aviation Itinerant Operations Forecast [14]

casts mostly followed a linear trend. As stated before, the forecasts are especially optimistic for commercial operations, as compared to expected growth in GA.

For all categories, the forecasting error increases with time, confirming a priori expectations of forecasting behavior. The commercial forecast slightly performed better than the forecast for itinerant GA operations for the first two years, but then for the second two, the itinerant GA forecast was closer to observed values. From the perspective of strictly measuring the forecast error as compared to reported data, the

Figure 2-6: 2006 General Aviation Local Operations Forecast [14]

forecast for local operations appears to have outperformed the other two.

This perceived improved accuracy in the local operations forecast, however, may

not be what it initially appears to be. While the 2006 and 2010 TAF data for

historical values overlap for commercial activity, the curves are disjoint for itinerant GA operations data, and differ even further for the local GA operations. This is a symptom of the data reliability; as discussed previously, data for towered airports is more accurate than the estimates given for non-towered airports and commercial activity tends towards towered airports while local general aviation tends towards smaller, non-towered airports.

An inspection of historical data and the forecasts reported in the TAF for non-towered airports shows that a large majority of airports have a constant level of operations across past and future years. Not only did they have zero growth as-sumed, the historical values show no change for the years 2001-2005. This indicates a tendency in reporting to mirror previous years, predisposing the forecast to greater apparent accuracy.

2.2.2

Towered Airports

Because of the issues of reliability of forecast accuracy for non-towered airports, a subset of 515 airports was chosen for analysis of the accuracies of the forecasts. These airports are in the NPIAS and also have operations data available from the FAA’s Air Traffic Activity System (ATADS) [4]. The forecasts of these towered airports are compared to the reported operations for fiscal years 2007-2011.

Figures 2-7, 2-8, and 2-9 compare the forecasts for towered commercial operations, itinerant GA operations, and local GA operations with their observed levels.

Figure 2-7: 2006 Towered Commercial Operations Forecast [14][4]

Figure 2-8: 2006 Towered General Avia-tion Itinerant OperaAvia-tions Forecast [14][4]

Figure 2-9: 2006 Towered General Avia-tion Local OperaAvia-tions Forecast [14][4]

The accuracy of the towered commercial operations forecast significantly improved as compared to the aggregate forecast. Conversely, the general aviation forecasts

significantly deteriorated. This difference in forecasting accuracy demonstrates a

very clear difference in the 2006 TAF’s forecast of commercial and GA operations. It appears that the forecasting effort put into commercial activity, especially at the nation’s biggest hubs, are more successful than that put into GA forecasts. The difficulty in forecasting GA activity may be due to the complexity of the drivers of general aviation.

Even though there has been a general trend downwards, starting even before 2006, the forecasts for much of the general aviation operations did not reflect a belief in the continuation of those trends. The forecast for general aviation is an optimistic one. The commercial forecast did perform much better in terms of short term forecast accuracy. Like the GA forecast, it is optimistic, but unlike the GA forecast, the optimism is more justifiable given the historical trend.

2.3

2001 - 2011 Aerospace Forecasts

Now, we shift our focus from evaluating the accuracies of general aviation forecasts to characterizing past and current forecasts, as reported in the FAA Aerospace Forecast.

2.3.1

Comparison of Past Forecasts

Figures 2-10 and 2-11 plot the observed and forecasted levels of activity for local and itinerant GA operations, respectively. Observed operations are included from 1994 to 2010, along with the the long term forecasts from the years 2000 to 2011 (note that the length of the forecasts have increased from 10 to 20 years across that time period). Clearly, a generalization can be made about these forecasts that the outlook on general operations are positive. All forecast years project nearly steady growth without any periods of decline, even when the observed trend for the previous years are negative. While there has been a continual optimism, the projected growth rates have slowly decreased from the 2000 to 2011 forecasts in response to the observed trend.

Figure 2-10: Local General Aviation Operations at All Towered Airports [15]

2.3.2

2011-2031 Forecast

Figures 2-10 and 2-11 show the forecast levels of local and itinerant operations until 2031. Local GA operations and itinerant GA operations are forecast to grow over the 20 year period by 27.1% and 28.2%, respectively.

The forecast posits that the general aviation fleet will increase from 224,172 air-craft in 2010 to 270,920 in 2031, growing at an average rate of 0.9% a year. Within

Figure 2-11: Itinerant General Aviation Operations at All Towered Airports [15]

this aggregate growth, the fixed-wing turbine aircraft fleet will grow at a rate of 3.1% per year, the fixed-wing piston aircraft fleet will grow at a rate of 0.2% per year, and the rotorcraft fleet will grow at a rate of 2.6% per year.

General aviation hours flown are forecast to increase from 24.1M in 2010 to 37.8M in 2031, an average annual growth rate of 2.2% a year. Forecasts for hours flown by aircraft type include fixed-wing turbine aircraft hours flown growing at a rate of 4.0% per year, fixed-wing piston aircraft hours flown growing at a rate of 0.7% per year, and rotorcraft hours flown growing at a rate of 3.0% per year.

2.4

Summary

Aviation forecasting is an imperfect science, and whether considering commercial or general or any other category of aviation, forecasting is a challenge. Both GA and commercial activities, driven by differing drivers, are dependent on conditions that are difficult to predict.

The 2006 TAF gives an indication that general aviation is either difficult to predict or is not given the amount of thorough formulation that commercial activity forecast-ing is given, yet GA plays a major role in the United States. While forecastforecast-ing is only one part of the planning or evaluations by its stakeholders, it does give some weight to the importance of GA forecasts in relation to its more accurately projected commercial counterpart. While it could be simply be said that the ”forecast is always wrong”, more can always be done to improve forecasting methods. Looking at the past trends and better identifying what stimulates or hinders general aviation activ-ity, in the context of the past and in the future may help better formulate forecasts

in the future.

In the past decade, general aviation forecasting work by the FAA has steadily remained positive, despite the almost consistent decline in activity, with the 2011 forecast being no different. In the next chapter, an examination of the current and past trends in America’s general aviation activity and its drivers will put into per-spective this optimistic outlook.

Chapter 3

Trends

A better understanding of past trends is necessary for forming better expectations of the future. This chapter takes a step back and explores some of the trends in activity over the past decades. Historical trends in fuel, socio-economic factors, pilot population, aircraft population, safety issues are also presented as factors that affect activity across the country.

3.1

Activity Trends

Some direct measures of general aviation activity in the United States include number of local and itinerant operations and hours flown.

3.1.1

Operations Trends

As discussed in Chapter 2, data on general aviation operations is most accurate and available for towered airports in the NPIAS. All trends presented in this section include only those related to towered general aviation operations, which are collected and published in the FAA’s Air Traffic Activity Data System (ATADS) [4].

Figure 3-1 plots towered general aviation operations across the United States from 1969 to 2010. Trends for local operations, VFR itinerant operations, IFR itinerant operations, total itinerant (VFR and IFR) operations, and total (local and total itinerant) operations are plotted in the figure. Note that only the total itinerant data is available before 1991 and a breakdown of this total into VRF and IFR operations is available after that year.

In the decade between 1969 and 1979, general aviation operations grew at a rapid rate, with total operations doubling in the short time span. The majority of this growth resulted from an increase in itinerant operations in this time span. After this initial ramp up period, both local and itinerant operations followed similar aggregate trends.

After the decade of growth, general aviation operations dropped dramatically by 32% from peak levels in 1979 to 1982, just in three years. This sharp decline is the most drastic change that general aviation experienced in the past 4 decades.

Figure 3-1: Towered Operations[4]

After that decline, operations began to climb again, albeit at a rate slower than that in the 1970s. Operations slowed again for five years beginning in 1991, before increasing again to the peak levels of the late 1970s in 2000.

Beginning in 2000, total operations have experienced their longest period of histor-ical decline. Between 2000 and and 2010, total towered operations across the country had monotonically decreased by 35%. Within this declining trend in recent years, a significant drop occurred between 2007 and 2009.

3.1.2

Hours Flown Trends

Another source of activity measures comes from the FAA’s General Aviation and Part 135 Activity Survey which reports estimates of total hours flown in general aviation operations every year that the survey has been administered. Figure 3-2 shows how hours flown by category of use has changed throughout this time. The downward trends found in towered operations are also echoed in the estimates of hours flown.

3.1.3

Spatial Trends

Figure 3-3 gives an indication of how those total operations in recent years have been distributed across the country. The top 50 general aviation airports, with respect to total GA operations, are shown on the map of the United States, with the size of the markers relating to the number of operations at the respective airports. The map is also overlayed with average annual precipitation in the United States for the years 1961-1990, where the red end of the color spectrum indicates low levels of precipitation and the purple end of the spectrum indicating high levels of precipitation.

These 50 airports account for 33% of all towered general aviation operations in the country. The airports also occur in clusters, indicating a link between spatial locations and activity levels. A strong concentration of operations are in the southwest, where average rainfall is at a minimum in the country. The airports are generally clustered in the southern half of the country,in Southern California, Arizona, and Florida where unfavorable winter conditions aren’t as prevalent.

3.2

Drivers and Related Trends

3.2.1

Pilot Population

The pilot population in the United States, shown in Figure 3-4, has declined since its peak in the 1980s across all categories of certification. It decreased by 28% from a peak of 827,071 certificates in 1980 to 594,285 in 2009. An increase in total certifi-cates occurred in 2010 and it could partially be the result of an increase in student certificates due to the validity of those certificates being extended by the FAA to 60 months.

Figure 3-4: Pilot Certificates Held by Category [16]

The decline is even more evident in Figure 3-5, which considers the number of certificates as a proportion of the U.S. population in a given year. This percentage dropped from 0.36% in 1970 to 0.20% in 2010.

Figure 3-5: Percentage of Pilot Certificates Held in the U.S. Population, by Category [16]

This decline in population indicates that pilots are retiring at a rate higher than the rate at which student pilots are beginning to fly and become certificated. This trend can be seen in the data collected from the FAA’s General Aviation and Part 135 Activity Survey. Figure 3-6 shows how the number of instructional flight hours has decreased by a significant 53% from 1990 levels.

Figure 3-6: Instructional Hours [2]

As could also be expected from the slow addition of new pilots into the pilot pop-ulation of the country, the average age of those left in the poppop-ulation are increasing. Figure 3-7 shows how the average age of pilots has steadily increased from 37.8 years in 1981 to 44.2 years in 2010.

Figure 3-7: Average Age of Active Pilots by Category [16]

3.2.2

Aircraft Population

Along with the decline in pilot population, there has been a decrease in the manu-facturing of general aviation aircraft in the United States and an over all aging of the aircraft fleet.

Figure 3-8 shows the numbers of general aviation aircraft shipments by US man-ufacturers from 1960 to 2010.

Figure 3-8: General Aviation Shipments by Type of Aircraft Manufactured in the United States [16]

General aviation aircraft manufacturing experienced a strong boom in the 1970s, which flooded the country’s active fleet. Figure 3-9 shows how this boom corresponds to the rapid growth in towered operations in the same time period. However, by the early 1980s, the aircraft manufactures were significantly hindered by rising insurance

costs brought on by product liability lawsuits as aircraft aged and resulting accidents were brought into litigation through tort law. From the peak of production in 1979 to its low in 1993, the number of shipments decreased by a staggering 94%.

Figure 3-9: Towered General Aviation Operations and Events Affecting Aircraft Pop-ulation

Shipments again began to grow after more than a decade of stagnation when the General Aviation Revitalization Act (GARA) was passed into law in 1994. The new statute limited the liability of aircraft manufacturers to accidents involving their aircraft in less than 18 years after the delivery of the aircraft [20]. This removed the crippling financial burden on aircraft manufactures and allowed for a more aircraft production. Also during this time, an increase in fractional ownership programs, benefiting from the favorable tax treatment for co-ownership, stimulated demand for aircraft. A dip in shipments observed from 2000 to 2003 is most likely a result of economic recession and the 9/11 terrorist attacks. Shipments resumed growth until 2008, until the economy was hit with another significant recession [23].

Although there has been a slow recovery of growth in the past two decades, the percentage of of general aircraft exported to foreign countries has increased from 20.3% in 1978 to 51.6% in 2010. Figure 3-10 adjusts the number of total aircraft shipments to account for those aircraft that are exported out of the country from 1978 to 2010. It is now clearer that of the small increases in shipments in the 2000s, only around half of those have stayed in the country.

Figure 3-10: General Aviation Total Shipments, Less Foreign Exports, by Type of Aircraft Manufactured in the United States [16]

The decrease in shipment levels are accompanied by an increase in billings as seen in Figure 3-11. Given the significant reduction in aircraft production, the increase in total billings suggest high increases in aircraft costs, even in constant dollars.

Figure 3-11: General Aviation Total Billings by Type of Aircraft Manufactured in the United States [16]

More expensive and fewer new aircraft entering lead to the domination of older

aircraft in the population. The resulting increase in maintenance costs and new

acquisition costs imply that operating costs in general aviation have been on the rise, which could contribute significantly to the decrease in activity.

3.2.3

Fuel

Fuel cost is one of the most basic, easily tracked, and understood drivers of aviation activity. As fuel prices increase or the volatility of prices increases, it should be expected that flying activity levels will be affected negatively.

Figure 3-12 shows how fuel consumption has changed since the 1980s. Aviation gasoline used to fuel piston type aircraft has experienced steadily decreasing sales over this time period, while jet fuel has increased.

Figure 3-12: Prime Supplier Sale Volumes [17]

Figure 3-13 shows how fuel prices have changed since 1978 in constant 2005 dollars. The shaded regions on the graph show periods of significant increase in fuel prices and they correspond to the shaded regions in Figure 3-14. During the periods of high fuel price increases, towered operations are in significant decline.

Figure 3-13: Fuel Sale Prices [18]

Figure 3-14: Towered General Aviation Operations and Rising Fuel Prices

Along with increased fuel prices, the availability of fuel threatens operations in the future. The future of the most widely used aviation gas used by piston aircraft, 100LL, is currently in question because of decreasing levels of demand limiting the available supply delivered to airports and because of environmental agencies working to eradicate the leaded fuel. In 2010, the U.S. Environmental Protection Agency

issued an Advance Notice of Proposed Rulemaking (ANPR) which announced an investigation into current emission engine standards. This action was in response to pressure from the environmental advocacy group Friends of Earth to regulate or eliminate the leaded gas. The group argued that half of all lead found in the atmosphere can be attributed to piston-engine aircraft emissions and that the public’s exposure to this lead may lead to serious heath effects, especially in children [19].

3.2.4

Socio-economic factors

As discussed in Chapter 1, the national GDP alone is not a strong indicator of ag-gregate general aviation activity in the country. Figure 3-15 and Figure 3-16 show the trends of other socio-economic factors, disposable personal income per capita, in chained 2005 dollars, and the unemployment rate in the United States, respectively.

Again, disposable personal income alone does not appear to be a good indicator of activity. Its positive and mostly steady trend does not track the patterns in activity levels. While increasing levels of disposable income are expected to correlate with higher levels of flying activity because of its relation to how much pilots can spend on flying, the trend alone does not take into account costs associated with flying and therefore cannot alone give an indication of activity.

Figure 3-15: Disposable Personal Income Per Capita Chained 2005 Dollars [21] Unemployment is hypothesized to affect activity levels in the country as lower levels of unemployment indicate better economic conditions for businesses, more pilots involved in aviation related jobs, and more pilots or potential pilots being able to financially support their flying activities. The peaks of unemployment between 1970 and 1985 did in fact coincide with periods of increase in towered GA activity. The most recent spike in unemployment, from 2008 and 2010 related to a strong economic recession, coincides with the begin of the steeper drop-off in activity beginning in 2008. The beginning of the two declines in activity starting in 1991 and in 2000 also coincides with two periods economic recessions in the country, as seen in Table 3.1. Figure 3-17 illustrates how periods of economic recession, shaded green, relate to periods of decline in activity.

Figure 3-16: Unemployment [22]

Table 3.1: Economic Recessions in the United States [23]

Recession Length Peak Unemployment Rate GDP Decline

July 1990 March 1991 8 months 7.8% -1.4%

March 2001 - November 2001 8 months 6.3% -0.3%

December 2007 - June 2009 1 year, 6 months 10.0% -5.1%

3.2.5

World Events

As also noted in Figure 3-18, downturns in aggregate general aviation levels have also coincided with events occurring in the United States, other than those directly related to economics or the aircraft production rates discussed previously.

Figure 3-18: Towered General Aviation Operations and World Events

Along with the economic recession of the early 1990s, the decline in activity in the first half of the 1990s may also be linked to the country’s involvement in the Persian Gulf War (August 1990 February 1991) and to Hurricane Andrew (August 1992), the costliest natural disaster in Florida’s history that destroyed airport facilities and aircraft in one of the nation’s most active area of general activity [24].

The current decline in activity that began in the early 2000s that coincided with economic downturn was further compounded by the terror attacks of September 11,2001. The attack resulted in the immediate halt of all activity for days and its more prolonged effects came in the form of poor perceptions of aviation in the public, increased security procedures at airports, and tighter regulation of the airspace [25].

3.2.6

Safety

As with any type of aviation, safety or perceived safety is an issue that can affect how the public, users, or potential student pilots view general aviation and, therefore, affect activity.

The number of fatal and non-fatal general aviation accidents in the United States along with the number of accidents per 100,000 flight hours, as reported by the Bureau of Transportation Statistics, are presented in Figure 3-19. Since the late 1960s, the number of general aviation accidents has been on an almost steady decline. An even more dramatic trend is seen in the number of accidents per 100,000 flight hours, which

has followed a near exponential decline since 1938. It dropped from over 30 accidents per 100,000 flight hours in the late 1960s to 5 accidents per 100,000 flight hours in 2010.

Figure 3-19: General Aviation Accidents [26]

Figure 3-20 shows that the absolute number of fatal accidents has been generally

decreasing since the beginning of the 1980s. A decrease of 1,000 fatal accidents

occured over the past two decades. The total fatalities per accident, however, has not dropped at the same rate as the number of accidents and is quite less steady, fluctuating between 0.4 and 0.3 fatalities per fatal accident.

Accidents and accident rates have been on the decline for the past 50 years, while total fatalities per accident have stagnated in recent years. There is no strong correlation between the measures of activity discussed in the first section of this chapter and the measures of general aviation safety. The more positive indication of an advancement in safety does not seem to be a primary driver of higher aggregate activity levels.

3.3

Conclusions

Despite the optimism in the forecasts seen in Chapter 2, general aviation activity, whether measured by local operations, itinerant operations or by hours flown, has dropped since its peak in the late 1970’s. There have been periods of growth and decline in activity since then that allowed activity to reach those levels again by end of the 1990s, but a steady decline has persisted for next two decades.

General aviation activity grew to peak levels in the 1970s, accompanied by high production rates of new general aviation aircraft, until it was set back by liability issues faced at the end of the decade and volatile fuel costs. Operations levels slowly recovered, until the 1990s, when war, natural disasters, and economic downturn, resulted in decline. Activity was only allowed to ramp back up for five years be-fore economic downturn marked the inflection point in the trend, which was further compounded by the 9/11 terrorist attacks, rising fuel prices throughout most of the decade, and world wide economic recession at the end of the decade. Concurrently, there has been a decline in the pilot and aircraft populations, with less student pilots, on average an older pilot population, more expensive new aircraft, and an older active aircraft population.

The drivers and related trends explored in this chapter work in both independent and confounding ways to produce the levels of activity realized. While however some major factors affecting activity presented themselves, they are also by no means the only factors at play. In Chapter 4, a survey of general aviation pilots is presented as a way to further clarify these complicated relationships and to shed more light on what affects an individual’s flying activity and on what they believe is in store for the future of general aviation.

Chapter 4

Survey

The data presented in the previous chapter gives insight into aggregate trends in general aviation across the country. While major drivers or related trends presented themselves in this data, we seek to gather a deeper understanding of what factors affect a given pilot’s flying activity on an individual level. To help achieve this goal, an exploratory survey of general aviation pilots was employed to gather the experiences and opinions of a sample of active and inactive pilots.

The chapter begins with a discussion of the motivation behind using a survey to collect more data on general aviation activity. The design and sampling method of the survey are then presented. This is followed by a discussion of the methodological issues facing this design and any implications that it may have on the analysis of the survey’s results. Then a profile of the survey respondents is presented, along with discussion of how well it fits the population of interest. The chapter concludes with a brief commentary on the effectiveness of the survey in achieving its methodological goals.

4.1

Goals

An important step in understanding the state of general aviation in the U.S. is to characterize it numerically, through the analysis of the trends in easily quantified data, as was done in the previous chapter. However, subtleties in the drivers of an individual’s flying activity can be lost when aggregating information across the entire population of pilots that produce the activity in the country. Therefore, we look for other information that will provide another perspective that will either compliment the findings in the data or to provide new insight.

In order to compliment the current available data and to paint a more complete picture of the state of general aviation, an exploratory survey of general aviation pilots was developed and administered. A survey is used as an efficient tool in learning more about the opinions and experiences of a group of people which is too large to survey in full. The survey aims to gather information from a population as large as the hundreds of thousands of pilots in the United States, including those who may or may not be currently active.

The purpose of the survey is to two fold. The first is for information gathered from the survey to either confirm or dispute the hypothesized relationships presented in the previous chapter. The results can help to separate correlation from causation in the data analysed, making clear what in fact drives or hinders activity. In addition to this role, it is also aimed at obtaining information that are not directly reflected in the data. It asks for experiences, and in particular, opinions on the past, current, and future issues in general aviation.

Goals of the survey include answering questions such as:

• What are the experiences of pilots in terms of when they started flying, why they began flying, and what factors hindered or bolstered their flying activity throughout their flying careers.

• Are the experiences or opinions of GA pilots more or less the same across population groups or do they differ significantly? If they differ, do they change with factors such as a pilot’s age or where they live or fly?

• What do pilots expect in the future and what do they think are the biggest challenges facing general aviation?

The hope is that the information provided directly by GA users will add another dimension to the data analysis and will make the results of this research more useful to those who have an interest in GA’s future.

4.2

Design

4.2.1

Survey type

The survey is self administered survey hosted on the internet. The survey includes a combination of open ended, multiple choice, interval scale, and ratio scale questions.

Content

The pages of the survey, as seen by the participant, are in Appendix A.

• The survey begins with an explanation of what the survey is about, who is administering the survey, and what the goal of the survey is. The terms of par-ticipation and confidentiality, along with contact information for any questions regarding the survey are presented to the participant.

• The participants are then asked to best describe their current certification and ratings. Total flight time, age, gender, and occupation are also collected. This demographic information is used to gauge how well the survey respondents represent the pilot population of interest and also to group respondents by similar characteristics in the analysis of the findings.

• The participants are asked when they began flying and whether or not their activity increased or decreased since that time. The are also presented with a list of potential reasons a person might learn to fly and are asked to select the reasons that applied to themselves.

• The participants are then asked if they flew in 2011. If they did, they are asked to describe their activity in 2011 by providing hours flown, the state based from, the regions flown in, and the primary aircraft type flown.. If they did not fly in 2011, they are asked to describe why they did not fly in 2011 in an open response format.

• The participants are asked to identify the period of time, in 5 year increments, that they were most actively flying. If 2011 was the peak of their flying activity, are asked to describe why 2011 was their peak year of flying. If they choose any other period, they are asked to characterize their flying during their indicated peak year of flying by providing hours flown, the state based from, the regions flown in, and the primary aircraft type flown.

• If 2011 was not their peak flying period, the participants are asked to describe why their flying decreased from their peak year to 2011 in an open response format. They are then asked to indicate how significantly changes in different factors (including costs, income, employment, enjoyment, available time, etc.) contributed to their decline in flying hours.

• The participants are asked to select how much they agree or disagree with a set of given statements relating to general aviation issues. These statements were formulated to test the validity of a number of hypotheses of what affects flying activity that emerged from the data analysis in Chapter 3. In these questions, they are providing their opinion on issues including aviation careers, flying activity and economic issues, enjoyment of flying, the affect of the internet on flying, increased regulations, fuel costs, and new technologies.

• Finally, the participants are asked to characterize their outlook on their personal flying and on the future of general aviation in the U.S. They are asked how they expect their flying to change over the next few years and also if there is any fuel price that would significantly hinder their flying activity. Free response questions also ask what would stimulate their activity and what they believe to be the biggest challenges facing the general aviation community.

Best Practices

The survey was developed with the approval of MIT’s Institutional Review Board for adherence to best practices in social sciences research. In the goal of protecting the rights and welfare of the participants, the questions and survey methods were designed to be minimally invasive, with only enough detail to make meaningful conclusions. Along this vein, the participants answers were collected with SSL encryption, so that

responses are encrypted securely to the survey host’s servers. The participants were informed that no email addresses or IP addresses would be collected, that they would not be contacted afterwards, and that demographic information provided could not be used to identify the participants. They were also informed that their participation was voluntary, they would not be provided compensation for their participation, and that they could chose not to answer questions. The participants were provided with the email addresses of the survey administrators if they had any issues or concerns.

4.3

Sampling Method

Given cost and time constraints, a non-probability method of sampling was used for the survey presented here. To obtain data in a time frame acceptable to the project and within the constraints of the research budget, the method used was combination of methods defined as convenience and snowball sampling. Convenience sampling involves sampling those in the population who are readily available or are easy to access for the survey. Snowball sampling relies on people who have participated in the survey to refer the study to others who they seem fit to participate.

The survey was hosted online. Numerous general aviation groups and news sources were contacted about the research project and the survey, where they in turn informed their readers of the survey either through articles or newsletters. The biggest of these groups was AVweb, an independent aviation news source, with over 225,000 general aviation pilots and aircraft owners, OEM’s, suppliers and related businesses who subscribe to AVweb’s e-newsletters and visit the website. AVweb ran an sponsored advertisement in its two weekly e-newsletters for four weeks, which called for the input of general aviation pilots and was linked to survey. It is also assumed that some number of respondents were referred to the survey by other pilots who participated.

4.4

Sources of Error

In the previous section, it was discussed that the sampling method used was not an ideal random sampling method and therefore the prerequisites for utilizing most familiar statistical analysis relating the results of the survey to the total population are not met. The survey results are subject to greater sampling error, the amount of inaccuracy in estimating a value (average, proportion, etc.) intended to represent an entire population from the value obtained from a sample.

Inaccuracy occurs not only due to the fact that the sample is not random, but also on account of coverage, sampling, nonresponse, and measurement errors [27].

• Coverage Error: Coverage error occurs when every person in the population of interest does not have a known and non-zero chance of being included in the sample. The very design of this survey limits it to those who are familiar with using a computer and have access to the internet. The method of distribution of the survey also limits coverage to those who are subscribed to general aviation